Distortion Agnostic Deep Watermarking

TL;DR

This paper introduces a distortion-agnostic deep watermarking framework that leverages adversarial training and channel coding, enabling robust watermark embedding without explicit modeling of unknown or non-differentiable distortions.

Contribution

The proposed method removes the need for explicit distortion modeling during training, improving robustness to unknown distortions compared to existing deep learning-based watermarking techniques.

Findings

Achieves comparable results on known distortions

Outperforms existing methods on unknown distortions

Uses adversarial training and channel coding for robustness

Abstract

Watermarking is the process of embedding information into an image that can survive under distortions, while requiring the encoded image to have little or no perceptual difference from the original image. Recently, deep learning-based methods achieved impressive results in both visual quality and message payload under a wide variety of image distortions. However, these methods all require differentiable models for the image distortions at training time, and may generalize poorly to unknown distortions. This is undesirable since the types of distortions applied to watermarked images are usually unknown and non-differentiable. In this paper, we propose a new framework for distortion-agnostic watermarking, where the image distortion is not explicitly modeled during training. Instead, the robustness of our system comes from two sources: adversarial training and channel coding. Compared to training on a fixed set of distortions and noise levels, our method achieves comparable or better results on distortions available during training, and better performance on unknown distortions.